Defects in mismatch repair (MMR) have been shown to be associated with both Hereditary Non-Polyposis Colon Cancers (HNPCC) and sporadic cancers. These cancers have tumor specific mutations in microsatellite repeat sequences. These repeat sequences are thought to be prone to slipped strand mispairing during replication. If not corrected this will lead to replication errors and an accumulation of insertion/deletion (I/D) mutations. This phenotype has been termed MIN (Microsatellite Instability). Acquisition of the hypermutable phenotype leads to destabilization of the genome and appears to be an early event in cancer development. Hypermutable cancer cell lines fall into several distinct classes with regard to their mutation spectrum. The majority of these are MIN + and are mismatch repair deficient. Defective MMR function contributes to the hypermutable cancer phenotype. Some hypermutable cancer cell lines have microsatellite instability and are proficient in mismatch correction. That is, they recognize and correct all base-base mispairs and single stranded DNA loops up to nucleotides. This finding suggests other DNA repair pathways may be defective in these cells. Preliminary evidence shows that a pathway which repairs large DNA loops exists and that it is distinct from MMR. The hypothesis that this corrective pathway may also predispose to MIN will be tested directly in hypermutable cancer cell lines using DNA loop substrates. I will explore whether large loop repair is defective in these cell lines and whether it contributes to the MIN+ phenotype. To achieve this I plan to: 1. Screen hypermutable MIN cell lines for repair of large DNA laps. 2. Determine whether large DNA loop repair is distinct from mismatch repair. 3. Determine the functional requirements of this repair system. 4. Isolate the components of this novel repair system. This effort will initially focus on the DNA loop-specific recognition protein. A second project involves screening of a second class of hypermutable cancer cell lines which are proficient in MMR. These are hypermutable in an in-vitro assay which detects transition/transversion mutations at the HPRT locus, but do not demonstrate MIN. The mutation spectra of these cell lines is consistent with defects in repair systems, such as base excision repair, which corrects specific misincorporated bases or those produced spontaneously by hydrolysis. I will screen for defects in several glycosylases using a simple assay and further study defective cell lines.